Hockey Goals

Fill in a module description here

Imports

# this next line shouldn't have to be here





import warnings

import matplotlib.pyplot as plt
import pandas as pd
import seaborn as sns
from matplotlib.ticker import FuncFormatter
from pandas.plotting import register_matplotlib_converters
from ydata_profiling import ProfileReport

register_matplotlib_converters()
sns.set()
sns.set_context("notebook")
plt.rcParams["figure.figsize"] = 10, 6
pd.options.display.max_columns = None
pd.options.display.max_rows = None
pd.options.display.precision = 4
warnings.simplefilter(action="ignore", category=FutureWarning)

dollar_formatter = FuncFormatter(lambda x, pos: f"${x:,.0f}")
thousands_formatter = FuncFormatter(lambda x, pos: f"{x:,.0f}")

Constants

DAYS_IN_YEAR = 365.25

Data

Raw

game_goals_raw = pd.read_csv(
    "https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2020/2020-03-03/game_goals.csv"
)
top_250_raw = pd.read_csv(
    "https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2020/2020-03-03/top_250.csv"
)
season_goals_raw = pd.read_csv(
    "https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2020/2020-03-03/season_goals.csv"
)

game_goals_raw.head()

	player	season	rank	date	game_num	age	team	at	opp	location	outcome	goals	assists	points	plus_minus	penalty_min	goals_even	goals_powerplay	goals_short	goals_gamewinner	assists_even	assists_powerplay	assists_short	shots	shot_percent
0	Alex Ovechkin	2006	1	2005-10-05	1	20-018	WSH	NaN	CBJ	Home	W	2	0	2	1	2	1	1	0	0	NaN	NaN	NaN	5	40.0
1	Alex Ovechkin	2006	2	2005-10-07	2	20-020	WSH	NaN	ATL	Home	L	0	1	1	-2	0	0	0	0	0	NaN	NaN	NaN	1	0.0
2	Alex Ovechkin	2006	3	2005-10-08	3	20-021	WSH	@	ATL	Away	L	0	1	1	0	4	0	0	0	0	NaN	NaN	NaN	3	0.0
3	Alex Ovechkin	2006	4	2005-10-10	4	20-023	WSH	NaN	NYR	Home	W	1	0	1	1	2	0	1	0	1	NaN	NaN	NaN	6	16.7
4	Alex Ovechkin	2006	5	2005-10-12	5	20-025	WSH	@	CAR	Away	L	1	0	1	0	0	1	0	0	0	NaN	NaN	NaN	6	16.7

top_250_raw.head()

	raw_rank	player	years	total_goals	url_number	raw_link	link	active	yr_start
0	1.0	Wayne Gretzky	1979-99	894	1	/players/g/gretzwa01.html	https://www.hockey-reference.com/players/g/gre...	Retired	1979
1	2.0	Gordie Howe	1946-80	801	2	/players/h/howego01.html	https://www.hockey-reference.com/players/h/how...	Retired	1946
2	3.0	Jaromir Jagr	1990-18	766	3	/players/j/jagrja01.html	https://www.hockey-reference.com/players/j/jag...	Retired	1990
3	4.0	Brett Hull	1986-06	741	4	/players/h/hullbr01.html	https://www.hockey-reference.com/players/h/hul...	Retired	1986
4	5.0	Marcel Dionne	1971-89	731	5	/players/d/dionnma01.html	https://www.hockey-reference.com/players/d/dio...	Retired	1971

season_goals_raw.head()

	rank	position	hand	player	years	total_goals	status	yr_start	season	age	team	league	season_games	goals	assists	points	plus_minus	penalty_min	goals_even	goals_power_play	goals_short_handed	goals_game_winner	headshot
0	1	C	Left	Wayne Gretzky	1979-99	894	Retired	1979	1978-79	18	TOT	WHA	80	46	64	110	20.0	19	NaN	NaN	NaN	NaN	https://d9kjk42l7bfqz.cloudfront.net/req/20191...
1	1	C	Left	Wayne Gretzky	1979-99	894	Retired	1979	1978-79	18	INR	WHA	8	3	3	6	-3.0	0	3.0	0.0	0.0	NaN	https://d9kjk42l7bfqz.cloudfront.net/req/20191...
2	1	C	Left	Wayne Gretzky	1979-99	894	Retired	1979	1978-79	18	EDO	WHA	72	43	61	104	23.0	19	34.0	9.0	0.0	NaN	https://d9kjk42l7bfqz.cloudfront.net/req/20191...
3	1	C	Left	Wayne Gretzky	1979-99	894	Retired	1979	1979-80	19	EDM	NHL	79	51	86	137	14.0	21	37.0	13.0	1.0	6.0	https://d9kjk42l7bfqz.cloudfront.net/req/20191...
4	1	C	Left	Wayne Gretzky	1979-99	894	Retired	1979	1980-81	20	EDM	NHL	80	55	109	164	41.0	28	36.0	15.0	4.0	3.0	https://d9kjk42l7bfqz.cloudfront.net/req/20191...

Functions

def raw_game_to_clean(raw):
    return raw.assign(
        years_old=(
            (
                ((raw["age"].str.split("-").str[0].astype(float)) * DAYS_IN_YEAR)
                + raw["age"].str.split("-").str[1].astype(float)
            )
            / DAYS_IN_YEAR
        )
    )


def raw_top_to_clean(raw):
    # prob ffill raw_rank
    return raw


def raw_season_to_clean(raw):
    return raw


def list_of_top_n_playernames(top_n=8):
    return top_250.iloc[:top_n]["player"].to_list()

game_goals = raw_game_to_clean(game_goals_raw)
top_250 = raw_top_to_clean(top_250_raw)
season_goals = raw_season_to_clean(season_goals_raw)

EDA

game_profile = ProfileReport(game_goals, config_file="config_minimal.yaml")
top_profile = ProfileReport(top_250, config_file="config_minimal.yaml")
season_profile = ProfileReport(season_goals, config_file="config_minimal.yaml")

game_profile

<class 'ydata_profiling.profile_report.ProfileReport'>.__repr__ returned empty string

top_profile

<class 'ydata_profiling.profile_report.ProfileReport'>.__repr__ returned empty string

season_profile

<class 'ydata_profiling.profile_report.ProfileReport'>.__repr__ returned empty string

Plots

Warning – dataset might have problems – the sum of the goals in the seasons don’t match the top dataset?

for player_name in list_of_top_n_playernames(top_n=8):
    sum_goals = season_goals[season_goals["player"] == player_name]["goals"].sum()
    top_goals = top_250[top_250["player"] == player_name]["total_goals"].values[0]
    print(f"{top_goals - sum_goals}, {player_name}")

-115, Wayne Gretzky
-174, Gordie Howe
-64, Jaromir Jagr
-32, Brett Hull
-28, Marcel Dionne
-35, Phil Esposito
-139, Mike Gartner
0, Alex Ovechkin

What do the cumulative goals scored look like as a function of age for the top scorers?

# player_name = "Alex Ovechkin"
fig, ax = plt.subplots(figsize=(10, 6))
fig.patch.set_facecolor("w")

for player_name in list_of_top_n_playernames(top_n=8):
    plotdf = game_goals[game_goals["player"] == player_name]
    ax.plot(plotdf["years_old"], plotdf["goals"].cumsum(), label=player_name)

box = ax.get_position()
ax.set_position([box.x0, box.y0, box.width * 0.8, box.height])
ax.legend(loc="center left", bbox_to_anchor=(1, 0.5))
ax.set_title("Total Goals Scored by Age")
ax.set_xlabel("Age [Years]")
ax.set_ylabel("Goals Scored")
fig.tight_layout()

Clean up Notebook